Voice memo to text: what Apple’s transcript covers, and when it stops
Voice Memos on iPhone 12 and later already shows a transcript you can copy, in English, Spanish, Portuguese, Italian, French, German, Japanese, Korean, and Chinese. If your language or your recording falls outside that, save the memo and transcribe the .m4a with an on-device app instead: LoroNote covers 100+ languages offline, separates speakers, and exports SRT.
- Built-in transcript
- iPhone 12+, ten languages
- LoroNote
- 100+ languages, offline
- Memo format
- .m4a — imports directly
- Output
- TXT, or SRT with timecodes
- Price
- Free download · $39.99 lifetime
How to get a voice memo transcribed properly
- 1
Try Apple’s transcript first
Open Voice Memos, select the recording, tap More (···), and choose View Transcript or Copy Transcript. For a two-minute memo in a supported language, that is genuinely the whole job and nothing needs installing.
- 2
Save the memo to Files when you need more
From the memo’s share sheet, choose Save to Files. The recording is already a .m4a, so nothing is converted or re-encoded on the way out.
- 3
Import it into LoroNote
Tap Import File and pick the .m4a. The language is detected automatically from over 100 — including the ninety-plus that Apple’s built-in transcript does not cover.
- 4
Let it transcribe on the device
Whisper Large V3 Turbo runs locally. A one-hour memo takes about three minutes on an iPhone 15 Pro, with the network off if you prefer.
- 5
Name the speakers and export
If the memo caught a conversation rather than just you, the speakers are separated and labeled on the device. Rename them, then export TXT or an SRT that keeps both the labels and the timecodes.
Where Apple’s built-in transcript stops
The built-in feature is good and free, and for a lot of memos it is the right answer. These are the specific edges it does not cross:
- Languages: Apple documents ten. Whisper Large V3 Turbo handles over 100, with automatic detection and no setting to change
- Speakers: the built-in transcript is one continuous block with no notion of who is talking
- Subtitles: there is no SRT export, so nothing to hand an editor or upload as captions
- Devices and regions: transcription needs iPhone 12 or later, and Apple notes availability varies by country, region, and language
- Other people’s audio: the workflow is built around memos you recorded in the app, not a file someone sent you
How long it takes
Memos are usually short, so this mostly matters the one time you record a three-hour session by accident:
| Recording length | Time to transcribe |
|---|---|
| 5 minutes | about 16 seconds |
| 30 minutes | about 1.5 minutes |
| 1 hour | about 3 minutes |
| 3 hours | about 10 minutes |
Measured by LoroNote on an iPhone 15 Pro at roughly 19x real time. Newer devices transcribe faster; older supported models take longer.
Practical notes
No conversion needed — a memo is already .m4a
Voice Memos records AAC audio in an MP4 container. That is one of the formats LoroNote imports directly, so exporting and importing is a copy, not a re-encode.
Copy-and-paste stops working the moment you need to quote
Pasting Apple’s transcript into Notes is fine until you have to attribute a sentence or check it. There is no timecode to jump back to and no speaker label to attach the line to.
Long memos need battery, not patience
The app will record for hours. Before transcribing a long one, check free storage and charge, and leave the app in the foreground while it processes.
On-device transcription has a hardware floor
LoroNote needs iOS or iPadOS 18.6 or later and iPhone 13 or later, iPhone SE (3rd generation), or a supported recent iPad. Older hardware cannot run the model locally at all.
Voice memos are the most private audio on the phone
Look at what is actually in the app: the consultation you wanted to remember correctly, the conversation with a lawyer, the argument you recorded, the idea you would rather nobody read half-finished. It is the least considered and most sensitive audio most people own.
That is why the processing location is worth caring about here specifically. Apple’s transcript runs on the phone; LoroNote runs Whisper Large V3 Turbo on the phone too. Uploading the same memo to a web converter to get around a language limitation trades the one property that made it safe.
Common questions
Can I transcribe a voice memo without an internet connection?
Yes. LoroNote imports the .m4a and transcribes it on the device, so airplane mode changes nothing. Apple’s own Voice Memos transcript is also generated on supported devices rather than in a browser.
Does Apple’s Voice Memos transcribe my language?
Apple documents English, Spanish, Portuguese, Italian, French, German, Japanese, Korean, and Chinese (Simplified and Traditional) on iPhone 12 and later, with availability varying by country and region. Anything outside that list needs a different app.
How do I get speaker names onto a memo transcript?
Import the memo into an app with speaker identification. LoroNote separates the speakers on the device, labels every segment, and lets you rename "Speaker 1" once to update all of their lines at once.
Can I turn a voice memo into subtitles?
Yes. Export as SRT and you get a timecoded subtitle file — useful when the memo is the audio bed for a video, or when you simply want a transcript where every line is timestamped.
Will importing a memo delete the original recording?
No. Saving a memo to Files and importing it makes a copy; the recording stays in Voice Memos until you delete it there yourself.
Related
- M4A to text The format Voice Memos, Zoom, and QuickTime already save in — imported without conversion.
- Timestamped transcripts Sentence-level timecodes, tap-to-replay, and SRT export that keeps the timing.
- Subtitle generator (SRT) Timed .srt files from audio or video, with speaker labels, generated on the device.
Start with the memo Apple would not transcribe
Pick the one in the wrong language, or the one with three people talking, and see what comes back.
Download LoroNote